Six weeks in, and I still haven't touched a desktop. Everything architecture decisions, debugging, database migrations, the whole thing happens through a phone screen and an AI agent doing the actual typing.
What's strange is how different the failure modes are from what I expected. I thought the hard part would be not knowing how to code. It wasn't. The hard part was learning to think in prompts instead of logic how to describe a bug precisely enough that the agent fixes the real problem instead of patching the symptom, three times, in three different wrong places.
I keep wondering if this is a preview of something bigger, or just a novelty that works because I happen to have a lot of patience and a phone that doesn't die. Curious what others here think: for the people who've built with AI agents has it changed how you think about programming as a skill, or does it just feel like outsourcing typing?
And for anyone who hasn't tried building this way yet what's stopped you? Time, trust, or just never having a reason to?
Not meetings. Not interruptions. Just the work itself. The tasks I spent my day on.
60% of them were things someone else had decided needed doing. A request, a bug report, a question that turned into a task. None of it was on my list. All of it felt urgent in the moment.
Someone sent me a message walking me through a workflow they had built inside my product. Three tools chained together, a shortcut I forgot existed, a workaround for a limitation I had decided was permanent.
It worked. It was elegant. I had never considered using my own product that way.
I'll start with one I've often debated: "urgency". As in, fake countdown timers, or saying "you have 1 spot left" when there are infinite spots available.
The annoying part is that they work. Conversion rates go up by a few percentage points. So saying no isn't some clever long-game arbitrage where I secretly win in the end. It's just leaving money on the table because it makes me feel gross.
In the early days without a proper HR setup it s really hard to tell who genuinely believes in your vision versus candidates who just want a temporary gig while job hunting.
We had to change our interview style from standard technical coding tests to giving candidates messy real world problems to solve alongside us.
How do you test for true startup alignment and risk tolerance in early interviews before making an offer?
Making a product is not easy, and anxiety can be found in almost anything, what do you think is the part that gives you most anxiety when working on your product? Is there a method you use to combat that feeling, How do you deal with that? Let's Share
Four months of work. Design, code, testing, launch, promotion. The whole thing.
Usage never crossed 3%. The people who used it loved it. Everyone else ignored it. And it was making the rest of the product harder to understand because it added a whole section to the navigation that most people never clicked.
I built something fast, reliable, handled every case I could imagine. It also fell apart the moment two people used it at the same time. A race condition I had never hit because I was always the only person testing it.
The launch brought the second person. And the third. The feature that had been perfect for months broke within an hour.
When outbound campaigns underperform, the knee-jerk reaction is to check domain health, SPF/DKIM records, and technical setup. We spend days optimizing deliverability metrics.
But looking closely at open to reply ratios, the technical setup was fine. The issue was that the email sounded like an automated sales script disguised as a quick chat. Technical infrastructure gets your message to the inbox, but clear, humanized messaging is what keeps it out of the trash.
How do you audit your copy to ensure it sounds like a real human writing to another human, rather than a bot following a sequence?
